Sales forecasting

Title Sales forecasting
Authors Nazmul Alam Diptu(
Supervisor Dr. Nabeel Mohammed
Semester Fall, 2018

Time Series Analysis and Prediction is the driving force in modern Data Science advancement. Finding hidden patterns inside millions of data is opening up new areas of improvement in both business and research. Data Science Competitions create an open source environment among the data scientists, engineers and analysts. We attempted in one of those challenging competitions to put our knowledge on test. We are given a challenging time-series dataset consisting of daily sales data, kindly provided by one of the largest Russian software firms – 1C Company. Our job is to predict total sales for every product and store in the next month. The list of shops and products slightly changes every month. We are currently on 10th position among around 1900 teams. Our Leaderboard score is 0.87915 in Root Mean Square Root Metric .